ietf-dkim
[Top] [All Lists]

Re: [ietf-dkim] lets add one more shall we?

2007-06-08 07:17:30

On Jun 8, 2007, at 1:59 AM, Charles Lindsey wrote:

On Thu, 07 Jun 2007 14:41:22 +0100, Hector Santos <hsantos(_at_)santronics(_dot_)com> wrote:

Charles Lindsey wrote:

In other words, not all MX query gives you IP addresses to try. So in that case, you can have an MX that directs you to the domain nomail.invalid (which has no A record, of course) and that is the end of the matter. What is wrong with that?

A change in long establish SMTP semantics and sending strategies.

No. Just a way to say this domain does not want to receive any email, even though it has an A record. That MX should get cached (so no excessive load on the authoritative server). And nomail.invalid should also get cached, as a DNS failure, or so I have been informed, so no excessive load on the root servers; a smart DNS resolver will already have built into it that 'invalid' is not even worth looking up. I proposed it as an alternative to the suggested "MX .", which apparently had problems.

This is a different name with a worse problem. An MTA is unlikely to know that either "." or "nomail.invalid" is an invalid host name. The "." is valid and represents "root". The "nomail.invalid" will cause a flood of unterminated transactions requesting name servers for "invalid". Perhaps this scheme could use "nomail.arpa" where there is a long lived record published to curtail the traffic that older MTAs will create. However a method that returns an "invalid-by- name" will cause these older MTAs to consider the hostname to exist when it does not.

-Doug
_______________________________________________
NOTE WELL: This list operates according to http://mipassoc.org/dkim/ietf-list-rules.html

<Prev in Thread] Current Thread [Next in Thread>